The overall goal of this longitudinal, randomized clinical trial (RCT) is to examine the bio-behavioral effects of mindfulness meditation (MM) in cancer patients undergoing stem cell/autologous bone marrow transplantation (SC/ABMT). SC/ABMT is a highly aggressive cancer treatment resulting in multiple significant side effects and lengthy hospitalization. The time in the hospital is both physically and emotionally challenging, while the weeks preceding and the months following are known to be quite stressful. Interventions to improve the SC/ABMT experience are warranted. MM has been shown to be helpful in persons with medical conditions, including cancer. However no published studies have investigated the use of MM in highly symptomatic, hospitalized patients. Preliminary work by this research team demonstrated that MM is feasible to carry out with hospitalized SC/ABMT patients and findings from the pilot study suggest benefits in alleviating treatment-related symptoms and inducing relaxation. The primary aim of this RCT is to evaluate the effects in a MM group, as compared to an attention control group and a usual care control group, on psychological functioning (perceived stress and distress), cancer-related physical symptoms, and overall quality of life over time (from before SC/ABMT and up to 6 months post-discharge). Secondary aims are 1) To compare changes in neuro-endocrine stress markers (24-hour urinary cortisol and catecholamine levels) and immune-related transplant markers (# days to bone marrow engraftment, length of BMT hospital stay, # infections, # days on antibiotic/antifungal therapy, and hospital readmissions post-BMT) among the three groups;2) To evaluate short-term changes in cardiovascular arousal (heart and respiratory rates) and symptoms and mood (visual analog scales) among the three groups;and 3) To explore associations among psychological, physical (symptoms), neuro-endocrine, immune-related, and patient background variables. A total of 280 SC/ABMT patients will be recruited from two clinical sites (estimate of final sample at 6 months =V210). Groups will be stratified by clinical site and transplant risk. Random effects modeling will model the primary outcomes over time and test whether the mean levels of these outcomes differ among the three randomized groups. If statistical evidence of differences is found, then the effects at each time point will be estimated. These findings could have valuable implications for oncology healthcare professionals related to the management of highly symptomatic and psychologically stressed, hospitalized cancer patients undergoing intensive procedures such as SC/ABMT.